Poster No:
826
Submission Type:
Abstract Submission
Authors:
Anastasia Dimakou1, Ilaria Mazzonetto2, Andrea Zangrossi3, Miriam Celli4, Davide Nuzzi5, Giovanni Pezzulo5, Maurizio Corbetta6
Institutions:
1Padova Neuroscience Center (PNC), Veneto Institute of Molecular Medicine, Padova, Italy, 2Veneto Institute of Molecular Medicine (VIMM), Padova, Italy, 3University of Padova, General Psychology, Padova Neuroscience Center, Padova, Italy, 4University of Padova, Department of Neuroscience, Padova, Italy, 5Institute of Cognitive Sciences and Technologies, National Research Council, Rome, Italy, Rome, Italy, 6Univesity of Padova, Department of Neuroscience,PNC,VIMM, Padova, Italy
First Author:
Anastasia Dimakou
Padova Neuroscience Center (PNC), Veneto Institute of Molecular Medicine
Padova, Italy
Co-Author(s):
Andrea Zangrossi
University of Padova, General Psychology, Padova Neuroscience Center
Padova, Italy
Miriam Celli
University of Padova, Department of Neuroscience
Padova, Italy
Davide Nuzzi
Institute of Cognitive Sciences and Technologies, National Research Council, Rome, Italy
Rome, Italy
Giovanni Pezzulo
Institute of Cognitive Sciences and Technologies, National Research Council, Rome, Italy
Rome, Italy
Maurizio Corbetta
Univesity of Padova, Department of Neuroscience,PNC,VIMM
Padova, Italy
Introduction:
The brain's most remarkable functions -imagination, problem-solving, memory, learning, and planning-often emerge when disengaged from the external world. During these moments, spontaneous neural activity replays representational states, supporting an internal world model (Pezzulo et al., 2021). Initially seen as a memory consolidation mechanism (Ji & Wilson, 2007), replay also predicts future paths, integrates past experiences, and reverses event sequences (Foster, 2017). It extends beyond the hippocampus to large-scale networks like the default mode and dorsal attention networks (Kaefer et al., 2022; Zhang et al., 2023) and supports tasks like planning and abstract learning in humans (Kurth-Nelson et al., 2016; Liu et al., 2019). Theories suggest replay aids offline consolidation or online cognition (Laumann & Snyder, 2021). Current computational models view it as generative, recombining experiences through a world model rather than exact memory (Stoianov et al., 2022), though its precise roles remain unclear.
Methods:
We examined how neural replay supports online learning and post-learning consolidation using high-density electrophysiology and a sequential learning task (N=40). Participants viewed images from five categories (face, scene, body, tool, scrambled) in an image categorization task. Images could be presented as a sequence (e.g., face > scene > body > tool > scrambled) or were presented randomly. The probability of sequence presentation varied across different blocks.Resting-state data were recorded before the task (10 minutes), after each task block (1 minute), and post-task (10 minutes).
Lasso-regularized logistic regression models were trained on task representations from 256-channel activations sampled at 100 Hz. Models were trained and tested on time points from -0.5s to 0.5s relative to image onset using 5-fold cross-validation and L1 regularization to prevent overfitting. Beta estimates and classifier prediction performance (~200ms post-image) were used to decode task representations during rest. Sequential reactivation of task-related states during rest was identified using temporally delayed linear modeling (Liu et al., 2021). Pairwise transitions between reactivated states were computed from a decoded state-space matrix and modeled with multilinear regression for lags (Δt) of 10–500 ms. Replay significance was assessed with permutation tests, using shuffled state identities to form null distributions. Replay strength was significant if it exceeded 95% of null peaks (FWE < 0.05).
Significant replay segments were analyzed for time-frequency power modulations (5–150 Hz) with cluster-based permutation tests. Behavioral data linked reaction times to sequence probabilities, with learning assessed via generalized linear modeling. Participants were classified as high or low learners based on median-split learning slopes.
Results:
Participants learned the sequence rule, showed faster Reaction Times (RTs) as the probability of sequence presentation increased. High learners, with steeper learning slopes, showed forward replay compressed to 40ms lag during the task, which increased early and decreased as the task progressed. Low learners exhibited offline reverse replay with slower lag (50ms), stable across the post-task rest period. On-task replay occurred during high-frequency activity (>90Hz) localized to occipital and parietal sensors and correlated with learning. Offline replay displayed broader frequency signatures (20Hz– >90Hz) and topographies, starting in motor sensors and spreading toward frontal, parietal, and occipital regions.
Conclusions:
Our results are in line with the idea that spontaneous replay supports online learning and offline consolidation. Task-specific replay occurs as compressed, high-frequency event facilitating learning, while post-task replay is slower, with broader frequency and spatial signatures. These findings highlight state-specific roles of replay in generative processes underlying online and offline cognition.
Learning and Memory:
Implicit Memory 1
Modeling and Analysis Methods:
Classification and Predictive Modeling
Multivariate Approaches
Task-Independent and Resting-State Analysis 2
Novel Imaging Acquisition Methods:
EEG
Keywords:
Multivariate
Other - Human Replay ; high-density EEG ; Sequence Learning ; Spontaneous Activity
1|2Indicates the priority used for review
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
I accept
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information.
Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
I do not want to participate in the reproducibility challenge.
Please indicate below if your study was a "resting state" or "task-activation” study.
Resting state
Task-activation
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Healthy subjects
Was this research conducted in the United States?
No
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel?
NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Yes
Were any animal research approved by the relevant IACUC or other animal research panel?
NOTE: Any animal studies without IACUC approval will be automatically rejected.
Not applicable
Please indicate which methods were used in your research:
EEG/ERP
Behavior
Provide references using APA citation style.
1) Foster, D. J. (2017). Replay Comes of Age. Annual Review of Neuroscience, 40, 581–602. https://doi.org/10.1146/annurev-neuro-072116-031538
2) Ji, D., & Wilson, M. A. (2007). Coordinated memory replay in the visual cortex and hippocampus during sleep. Nature Neuroscience, 10(1), 100–107. https://doi.org/10.1038/nn1825
3) Kaefer, K., Stella, F., McNaughton, B. L., & Battaglia, F. P. (2022). Replay, the default mode network and the cascaded memory systems model. Nature Reviews. Neuroscience, 23(10), 628–640. https://doi.org/10.1038/s41583-022-00620-6
4) Kurth-Nelson, Z., Economides, M., Dolan, R. J., & Dayan, P. (2016). Fast Sequences of Non-spatial State Representations in Humans. Neuron, 91(1), 194–204. https://doi.org/10.1016/j.neuron.2016.05.028
5) Laumann, T. O., & Snyder, A. Z. (2021). Brain activity is not only for thinking. Current Opinion in Behavioral Sciences, 40, 130–136. https://doi.org/10.1016/j.cobeha.2021.04.002
6) Liu, Y., Dolan, R. J., Higgins, C., Penagos, H., Woolrich, M. W., Ólafsdóttir, H. F., Barry, C., Kurth-Nelson, Z., & Behrens, T. E. (2021). Temporally delayed linear modelling (TDLM) measures replay in both animals and humans. eLife, 10, e66917. https://doi.org/10.7554/eLife.66917
7) Liu, Y., Dolan, R. J., Kurth-Nelson, Z., & Behrens, T. E. J. (2019). Human Replay Spontaneously Reorganizes Experience. Cell, 178(3), 640-652.e14. https://doi.org/10.1016/j.cell.2019.06.012
8) Pezzulo, G., Zorzi, M., & Corbetta, M. (2021). The secret life of predictive brains: What’s spontaneous activity for? Trends in Cognitive Sciences, 25(9), 730–743. https://doi.org/10.1016/j.tics.2021.05.007
9) Stoianov, I., Maisto, D., & Pezzulo, G. (2022). The hippocampal formation as a hierarchical generative model supporting generative replay and continual learning. Progress in Neurobiology, 217, 102329. https://doi.org/10.1016/j.pneurobio.2022.102329
10) Zhang, L., Pini, L., Kim, D., Shulman, G. L., & Corbetta, M. (2023). Spontaneous Activity Patterns in Human Attention Networks Code for Hand Movements. Journal of Neuroscience, 43(11), 1976–1986. https://doi.org/10.1523/JNEUROSCI.1601-22.2023
No